Post-fire vegetation recovery mapping using LiDAR and Synthetic Aperture Radar

<p dir="ltr">Global Ecosystem Dynamic Investigation (GEDI) data was used to estimate vegetation recovery for various burned areas in Blue Mountains National Park in Australia using airborne laser scanner (ALS) as reference data. The relationship between GEDI-derived canopy height and cover, and Sentinel-1 (S-1) Cross Ratio (CR) vegetation index was also investigated. A mean value of ALS canopy height and cover, and S-1 CR was estimated within each 25m buffered GEDI footprint. A linear model was used to regress GEDI metrics with suitable ALS and S-1 CR metrics. The results of the analysis in the low burned areas show a coefficient of determination (R²) of 0.55 and a Root Mean Squared Error (RMSE) of 6.1 for the relationship between GEDI and ALS canopy height metrics and an R² of 0.19 and RMSE of 0.23 for GEDI and ALS canopy cover metrics. The relationship between GEDI canopy height and S-1 CR has an R² of 0.092 and an RMSE of 14.6 while canopy cover and CR has an R² of 0.012 and an RMSE of 4.9. Similar patterns are observed for the analysis in the moderate, high, and extremely burned areas. This work demonstrates that GEDI products offer promising information for vegetation recovery mapping.</p>